My Autoscaled Instance Fails When Running "Git Clone" On A Private Repo. I

Answered

My autoscaled instance fails when running "git clone" on a private repo.

I do have the SSH key placed at /root/.ssh/id_rsa on the machine, and when I SSH into the machine and run sudo su; git clone <the repo> it succeeds.

Also, in the extra_vm_bash_script field: I added a whoami command which prints root , so it seems like the user being used to run the git clone during task execution is in fact root .

For context, here's the startup command that the autoscaler runs:

python -m clearml_agent --config-file /root/clearml.conf daemon --queue aws_4gpu_machines --docker python:3.9

Full log included...

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					BattyCrocodile47
				
					0
					 × 1

Votes Newest

Answers 34

Hi @<1541954607595393024:profile|BattyCrocodile47>

I

do

have the SSH key placed at

/root/.ssh/id_rsa

on the machine,

Notice that the .ssh folder is mounted from the host (EC2 / GCP) into the container,

'-v', '/tmp/clearml_agent.ssh.cbvchse1:/.ssh'

This is odd, why is it mounting it to /.ssh and not /root/.ssh ?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

Actually, dumb question: how do I set the setup script for a task?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					BattyCrocodile47
				
					0
					 × 1

Wow, it really does not want to show the output of those print statements in stdout. Here's the output of the task from the console after cloning it. Confirmed that the setup script and all code changes are present:

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					BattyCrocodile47
				
					0
					 × 1

Or the log of the init script?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					BattyCrocodile47
				
					0
					 × 1

Show more results

Write your answer

201K Views

34 Answers

2 years ago